Split lottery jackpot winners

My company does a number of bonding activities to try to promote staff unity. In addition to the usual happy hour and sports fantasy league participation, we also have a lottery pool that I administer. My criteria for selecting the drawings to participate in has always been pretty simple. I calculate the Expected Value of the Mega Millions and Powerball drawings, and we participate any time the tax-adjusted EV of the drawing with the jackpot's cash option exceeds .5. That has been my method with the Powerball for years, and I've done it with the Mega Millions since MN got that game in February.

My company recently hired a new finance guy, and he's been criticizing my method (over beers at happy hour, so nothing serious). He swears up and down that I must include a correction for the probability of a tied win. His point to me is that the chance of ties decreases the EV, and we are not meeting the .5 EV criterion as often as I claim. My counter argument to him is that the effect of the tie is only realized if we win and tie. To the extent that we have to win the freaking lottery in order for the tie to matter, I'm happy to have "incorrectly" played and then WON THE LOTTERY.

If you ran my office lottery pool, would you factor in the possibility of a split jackpot?

Let's begin with a possibly unwarranted assumption: Your EV criterion is reached when there are large jackpots. If so, then ties do matter. More people play any kind of lottery when the jackpot is large. More players mean larger odds of a tie.

I agree with the finance guy. Large jackpots induce more play, which increases the probability of a tie, and lowers the share of the jackpot you can expect to get if you win. I'd like to do a study on this, if I had the data, but I speculate that at some point the value of lottery ticket purchases starts to go down as the jackpot goes up at some point, due to increased competition. In my opinion that already happens with cash ball jackpots in bingo. I think the best opportunities are just before the jackpot is so large that it induces people that otherwise don't play to play.

We had a thread like this in April lottery expected value

BTW, can I interest you in my Rapid Lotto game?

http://wizardofvegas.com/forum/off-topic/gripes/2145-a-good-lotto-bet-really/2/

Just scroll down tot he second post on the page. Everyone's a guaranteed winner! In fact, in exchange for a pledge to buy 100 tickets, I'll throw in three, count'em three, jackpots at no extra cost!

I just did some searching and came up empty. What I'm looking for is a table that shows the jackpot amount and number of tickets sold for each drawing of any lottery. I would prefer Powerball or Mega Millions, but their web sites didn't seem to have this information. As I wrote in my last post, my hypothesis is that if the jackpot gets too big, the expected value starts to decrease, due to induced competition. It wouldn't surprise me if the optimal jackpot for Powerball and Mega Millions is $99,999,999, because lots of people won't play until the jackpots gets to 100M. If anyone can point me to such data I would be quite grateful.

I doubt you will find a published history table anywhere. My initial post in the April thread linked by miplet outlined my suggestion for how to come up with a good figure for the number of tickets the lottery people estimate will be sold for an individual drawing. I don't see a better way an individual could come up with an estimate.

Is this any help?

http://lottoreport.com/lottosales.htm

Well, dang! Nareed did indeed find a table.

I still think the method I suggested is the best one for getting a figure for sales for the upcoming drawing.

Quote: Doc

Well, dang! Nareed did indeed find a table.

I got lucky. All I did was copy what the Wizard said he was looking for and paste it on Google.